Lipase-Catalyzed Kinetic Resolution on Solid-Phase via a "Capture and Release" Strategy

Article abstract of DOI:10.1021/ja037922x

The lipase-catalyzed kinetic resolution of (R/S)-3-phenylbutyric acid 2 using solid-supported cyclohexane-1,3-dione (CHD) 6 is described. In each case the predominant enantiomer observed, after cleavage from the resin, was (R)-(?)-3-phenylbutyric acid (R)-2 (ee > 99%) rather than the expected (S)-enantiomer of 2. This observation is in contrast to the fact that Chromobacterium viscosum lipase shows high enantiospecificity for the (S)-enantiomer in the corresponding solution-phase hydrolysis reactions. The (R)-acyl group was subsequently released from the resin by NaOH hydrolysis, and the yield of the reaction could be improved by triple acylation of the resin. Copyright

Full text of DOI:10.1021/ja037922x

Published on Web 10/25/2003
Lipase-Catalyzed Kinetic Resolution on Solid-Phase via a
“Capture and Release” Strategy
Cara E. Humphrey,† Nicholas J. Turner,*,† Morag A. M. Easson,‡ Sabine L. Flitsch,† and
Rein V. Ulijn†
School of Chemistry, UniVersity of Edinburgh, King’s Buildings, West Mains Road, Edinburgh, UK, and
Organon Laboratories Ltd., Newhouse, Lanarkshire ML1 5SH, UK
Received August 14, 2003; E-mail: n.j.turner@ed.ac.uk
Scheme 1. CVL-Catalyzed Hydrolysis of Dimedone 1,3-Enol
Ester 1a
Lipases are industrially important enzymes that have found
widespread use in the resolution of chiral alcohols, amines and
carboxylic acids.1 In view of their diverse substrate specificity they
have also been used as reagents for the solution-phase parallel
synthesis of libraries of compounds based upon pharmacophoric
scaffolds.2 However, in comparison with proteases, little is known
about the activity of lipases toward resin-bound substrates,3 an area
of contemporary interest with applications in solid-phase synthesis4
including enzyme-cleavable linkers.5 Herein we report the first
example of a lipase-catalyzed acylation reaction on a resin-bound
substrate, the results of which may have applications in parallel
solid-phase synthesis and screening methodologies.
a Reagents and conditions: CVL, 10% MeCN/0.1 M KPi buffer (pH )
7.0), rt, 1.5 h (49%).
Scheme 2. Synthesis of CHD-PEGA1900 Resin 6 and
Lipase-Catalyzed Resolution Involving a “Capture and Release”
Strategya
We recently reported the synthesis of resin-bound cyclohexane-
1,3-dione (CHD) and demonstrated its use as a capture and release
reagent for the synthesis of amides.6 Furthermore, we have shown
that esters of cyclic-1,3-diketones are good substrates for lipases
and esterases7 and thus sought to combine these two ideas to
investigate lipase resolutions on resin-bound substrates. Screening
of a range of lipases and esterases against enol ester (R/S)-1 (Scheme
1) identified Chromobacterium Viscosum lipase (CVL) as possessing
good activity and high enantioselectivity [(E) ) 88],8 and hence
this enzyme was selected for initial studies. Previous work has
shown that enzyme-catalyzed reactions on solid-phase are greatly
enhanced by the use of PEGA1900 resin (commercially available
from Polymer Laboratories, Ltd.) which swells in aqueous solvents.9
The resin also has an open-pore structure allowing large enzyme
molecules access to the resin active sites.10
PEGA1900 supported CHD resin 6 was prepared as shown in
Scheme 2. Ethyl-3,5-dimethoxy-cyclohexa-2,5-dienecarboxylic acid
3, obtained by the Birch reduction and alkylation of 3,5-dimethoxy-
benzoic acid,11 was coupled (× 2) directly to PEGA1900 resin 4
using DIC in DCM/DMF (1:1) to give resin bound bis-enol ether
5. Acidic hydrolysis of 5, using a TFA/H2O/DMF (90:5:5) mixture,
yielded CHD resin 6. The loading of CHD resin 6 was determined
as 19 µmol/g (swollen) and 130 µmol/g (dry), by coupling with
(R/S)-3-phenylbutyryl chloride followed by standard sodium hy-
droxide release of the resin bound racemate.
a Reagents and conditions: (i) DIC, DCM/DMF (1:1), rt, 2.5 h × 2; (ii)
TFA/H2O/DMF (90:5:5), rt, 2.5 h; (iii) lipase, 0.1 M KPi buffer (pH )
7.0), rt, 16 h; (iv) 0.1 M NaOH, rt 1 h.
Table 1. Lipase-Catalyzed Resolution by “Capture and Release”
entry
ester
lipase
product
yielda
ee
1
2
3
4
5
6
7
8
(R/S)-7
(R/S)-8
(R)-8b
CVL
CVL
CVL
CVL
PCL
(R)-2
(R)-2
(R)-2
(R)-2
(R)-2
(R)-2
(R)-2
(R)-2
74
38
92
7
77
83
78
86
65
>99
>99
>99
59
62
>99
>99
(S)-8c
(R/S)-8
(R/S)-8
(R/S)-8
(R/S)-8
PPL
CVLd
CVLe
PEGA1900 supported CHD 6 was subjected to CVL-catalyzed
acylation using (R/S)-methyl 3-phenylbutanoate 7 and (R/S)-vinyl
3-phenylbutanoate 8 (Scheme 2) to yield resin-bound 1,3-enol ester
9. As is typical for solid-phase reactions, a large excess (ap-
proximately 10 equiv) of acylating agent was used to ensure high
conversions. The reaction was performed in aqueous potassium
phosphate buffer at neutral pH in the absence of organic solvent.
Under these conditions, the concentration of hydrophobic acylating
agent at the reaction site, within the PEGA1900 microenvironment,
is high.12 After overnight reaction, the resin was extensively washed,
a Calculated by comparison of HPLC peak areas with a control [NaOH
hydrolysis of chemically synthesized racemic enol ester]. b ee ) 96%. c ee
) 92%. d Double acylation. e Triple acylation.
followed by acyl group release from the resin (aqueous NaOH) to
give 3-phenylbutyric acid 2 which was analyzed by HPLC
(Chiracel-ODH column) to determine the yield and enantiomeric
excess (ee) of the reaction (Table 1). In each case the predominant
enantiomer observed was the (R)-acid 2 rather than the expected
(S)-acid despite the known preference for CVL to catalyze
hydrolysis of the (S)-enantiomer in solution (Scheme 1). Similar
results were obtained with Pseudomonas cepacia lipase (PCL) and
† University of Edinburgh.
‡ Organon Laboratories Ltd.
9
13952
J. AM. CHEM. SOC. 2003, 125, 13952-13953
10.1021/ja037922x CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
rate of hydrolysis of vinyl versus methyl ester. Previous unpublished
studies in our laboratories have shown that diffusion of enzymes
into PEGA1900 resin occurs over a period of ca. 1 h. When CVL
was allowed to preequilibrate with the resin for 1 h before addition
of vinyl ester 8, the ee of the liberated (R)-acid 2 decreased to
90%. Under these conditions the (S)-vinyl ester 8 presumably
undergoes a small degree of transesterification onto the CHD resin,
prior to hydrolysis. Longer preequilibration times did not lead to
further reductions in ee.
In conclusion, we have demonstrated for the first time that the
lipase-catalyzed kinetic resolution of racemic esters can be carried
on solid-phase using an acylation/deacylation capture and release
strategy. The reactions exhibit enantiospecificity which is consistent
with the operation of a parallel kinetic resolution process. Such
resin-mediated reactions should be amenable to automation,
particularly using parallel synthetic approaches, and in view of the
broad substrate specificity of available lipases they can be exploited
for the preparation of combinatorial libraries suitable for screening
against biological targets of interest.
Figure 1. Explanation of enantiospecificity of reaction.
porcine pancreatic lipase (PPL), both of which are also known to
be (S)-selective. Use of the methyl ester 7 (Table 1, entry 1) resulted
in a reasonable yield of (R)-acid 2 with modest ee, whereas the
more reactive vinyl ester 8 (Table 1, entry 2) gave a lower yield
but excellent ee. Subjecting the resin to a second acylation reaction,
after washing, resulted in an increased yield (78%) (Table 1, entry
7) which could be further enhanced (86%) with a triple acylation
(Table 1, entry 8).
To probe this unexpected enantiospecificity, the acylation
reactions were repeated using enantiomerically enriched acyl donors.
When (R)-vinyl ester 8 (ee ) 96%) was used, the (R)-acid 2 was
obtained in high yield and ee (Table 1, entry 3). However, switching
to the enantiomerically enriched (S)-vinyl ester 8 (ee ) 92%) also
gave the (R)-acid 2 in low yield but high ee (Table 1, entry 4). In
this case the lipase selectively catalyses acylation of the minor
contaminant (R)-enantiomer, in the presence of excess (S)-enanti-
omer.
Acknowledgment. This investigation was generously supported
by funds provided by Organon Laboratories, Ltd., the Engineering
and Physical Sciences Research Council (EPSRC No. 00316713),
and the Wellcome Trust. We thank Dr. Ian Sadler for his help with
the MAS-probe NMR spectra of resin samples.
Supporting Information Available: Experimental preparations for
ester (R/S)-1, resins 5, 6, esters 7, 8, and standard procedures for lipase-
catalyzed hydrolysis and transesterifications (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Bornscheuer, U. T.; Kazlauskas, R. J. Hydrolases in Organic Synthesis:
Regio- and StereoselectiVe Biotransformations; Wiley-VCH: Weinheim,
1999.
(2) Secundo, F.; Carrea, G.; De Amici, M.; Di Ventimiglia, S. J.; Dordick, J.
S. Biotechnol. Bioeng. 2003, 81, 391; Michels, P. C.; Khmelnitsky, Y.
L.; Dordick, J. S.; Clark, D. S. TIBTECH 1998, 16, 210.
The reaction in the absence of lipase showed negligible levels
of formation of 3-phenylbutyric acid 2 by HPLC, eliminating the
possibility of background reaction between CHD resin 6 and vinyl
ester 8. Similarly, the corresponding solution-phase reactions, in
the absence of lipase, between dimedone and either vinyl ester 8
or 3-phenylbutyric acid 2 showed no evidence of conversion.
The enantiospecificity of the reaction can be rationalized by
considering the competing enzymatic hydrolysis of the (R)- and
(S)-enantiomers of 8 in a lipase-catalyzed parallel kinetic resolution
analogous to that reported by Rakels et al.13 (Figure 1).
(S)-Vinyl ester 8 undergoes rapid formation of an acyl-enzyme
intermediate followed by hydrolysis to form (S)-acid 2. Solution-
phase hydrolysis of vinyl ester 8 revealed that the (S)-enantiomer
is completely hydrolyzed in less than 15 min under these conditions.
The corresponding hydrolysis of the (R)-enantiomer is much slower
than that of the (S)-enantiomer allowing the enzyme intermediate
to undergo transesterification to form the resin bound (R)-enol ester
9. An alternative scenario, which we cannot exclude at present,
involves initial acylation by both (R)- and (S)-esters (k1S > k1R)
followed by rapid hydrolysis of the (S)-enantiomer (k2S > k2R), thus
releasing (S)-acid into solution with accumulation of the (R)-
enantiomer on the resin.14
(3) Nanda, S.; Bhaskar Rao, A.; Yadav, J. S. Tetrahedron Lett. 1999, 40,
5905.
(4) Ulijn, R. V.; Baragana, B.; Halling, P. J.; Flitsch, S. L. J. Am. Chem.
Soc. 2002, 124, 10988.
(5) For examples see: (a) Bo¨hm, G.; Dowden, J.; Rice, D. C.; Burgess, I.;
Pilard, J.-F.; Guilbert, B.; Haxton, A.; Hunter, R. C.; Turner, N. J.; Flitsch,
S. L. Tetrahedron Lett. 1998, 39, 3819. (b) Grether, U.; Waldmann, H.
Chem. Eur. J. 2001, 7, 959.
(6) Humphrey, C. E.; Easson, M. A. M.; Tierney, J. P.; Turner, N. J. Org.
Lett. 2003, 5, 849.
(7) Humphrey, C. E.; Easson, M. A. M.; Turner, N. J. ChemBioChem 2003.
Manuscript submitted for publication.
(8) Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc.
1982, 104, 7294.
(9) For the use of PEGA1900 resin in the kinetic resolution of an amino acid
via thermolysin-catalysed coupling onto a resin-bound amine see; Ulijn,
R. V.; Bisek, N.; Flitsch, S. L. Org. Biomol. Chem. 2003, 1, 621.
(10) Kress, J.; Zanaletti, R.; Amour, A.; Ladlow, M.; Frey, J. C.; Bradley, M.
Chem. Eur. J. 2002, 8, 3769.
(11) (a) Birch, A. J.; Slobbe, J. Aust. J. Chem. 1977, 30, 1045. (b) Liepa, A.
J.; Wilkie, J. S.; Winzenberg, K. N. Aust. J. Chem. 1989, 42, 1217.
(12) Ulijn, R. V.; Bisek, N.; Halling, P. J.; Flitsch, S. L. Org. Biomol. Chem.
2003, 1, 1277.
(13) Rakels, J. L. L.; Schneiders, A. G. M.; Straathof, A. J. J.; Heijnen, J. J.
Biocatal. Biotrans. 1996, 13, 179. For a review on parallel kinetic
resolution of racemic mixtures, including the use of enzymes see: Dehli,
J. R.; Gotor, V. Chem. Soc. ReV. 2002, 31, 365.
(14) Attempts to provide evidence for this scenario, by carrying out the CVL-
catalysed hydrolysis of CHD resin loaded with (R/S)-9, lead to very low
levels of cleavage of carboxylic acid from the resin.
The difference in ee for transesterification with methyl versus
vinyl ester (Table 1, entries 1 and 2) is consistent with the faster
JA037922X
9
J. AM. CHEM. SOC. VOL. 125, NO. 46, 2003 13953